Recent studies suggest that salivary lysozyme (Lz), lactoferrin (Lf), peroxidase (Spx) and secretory IgA (sIgA) may interact in a common antibacterial system. Interactions obeserved can be synergistic or antagonistic, and appear to vary with concentrations of the proteins involved. Those findings have direct implications for clinical studies of the effects of Lz, Lf, Spx, or sIgA on oral health and ecology. Previous clinical research has focused on individual antimicrobial proteins, and the results have generally been inconclusive. A possible reason may be that in vivo effects are the outcome of interaction between all components of the salivary antimicrobial system. A multiple protein approach therefore may be needed. Such studies will require a multivariate approach to data analysis. The proposed study will describe patterns of common variation of Lz, Lf, Spx and sIgA in stimulated parotid saliva from a large population sample. Its specific aims are 1) to identify major components of intercorrelation between Lz, Lf, Spx and sIgA (adjusted for flow rate and total protein); 2) to identify groups of subjects showing similar salivary profiles of Lz, Lf, Spx and sIgA (adjusted for flow rate and total protein); 3) to identify extraneous variables which may influence salivary levels of Lz, Lf, Spx and sIgA; 4) to provide statistically valid coefficients which can be used to classify unknown individuals according to salivary antimicrobial protein profile; and 5) to investigate the longitudinal stability of membership in groupings obtained. For objectives 1, 2 and 4, samples of stimulated parotid saliva from a randomly-selected cross-sectional sample of 256 university sophomores will be assayed for Lz, Lf, Spx, sIgA and total protein. Subjects will also fill out a health status questionnaire for objective 3. For objective 5, a random sample of 64 persons will be drawn from the cross-sectional pool, and results obtained from saliva samples taken two weeks apart will be compared. Statistical methods to be used include principle components analysis, cluster analysis, and discriminant analysis. Information obtained will contribute to understanding of the role of the salivary antimicrobial protein system in oral health and ecology. The long-range goal of the proposed study is to generate hypotheses concerning that role which may be tested in clinical investigations and in controlled experimental models.